【发布时间】：2016-12-08 03:57:41
【问题描述】：
我将 mysql 服务器从 5.7.14 降级到 5.6.32。
这样做之后，我假设我应该按照我遇到的说明运行 mysql_upgrade。
但是，为此，我需要提供我拥有的 root 用户/p。但是，当我尝试这样做时，它似乎不接受那些：
mysql_upgrade -u root -p
输入密码：
寻找“mysql”为：mysql
寻找“mysqlcheck”为：mysqlcheck
错误：获取服务器版本失败！可能是由于未经授权的访问。
致命错误：升级失败
所以我尝试通过创建一个类似以下内容的初始化文件来重置密码： 为'root'@'localhost'设置密码=密码（'abcd1234'）； 然后用 mysqld_safe --init-file=/home/root/mysql-init &
调用它但是这个好像不行，大概是因为mysql无法正确启动而抛出错误。
0802 12:50:45 mysqld_safe Starting mysqld daemon with databases from /var/lib/mysql
2016-08-02 12:50:46 0 [Warning] TIMESTAMP with implicit DEFAULT value is
2016-08-02 12:50:46 0 [Note] /usr/sbin/mysqld (mysqld 5.6.32) starting as process 2635 ...
2016-08-02 12:50:46 2635 [Note] Plugin 'FEDERATED' is disabled.
/usr/sbin/mysqld: Unknown storage engine 'InnoDB'
2016-08-02 12:50:46 2635 [ERROR] Can't open the mysql.plugin table. Please run mysql_upgrade to create it.
2016-08-02 12:50:46 2635 [Note] InnoDB: Using atomics to ref count buffer pool pages
2016-08-02 12:50:46 2635 [Note] InnoDB: The InnoDB memory heap is disabled
2016-08-02 12:50:46 2635 [Note] InnoDB: Mutexes and rw_locks use GCC atomic builtins
2016-08-02 12:50:46 2635 [Note] InnoDB: Memory barrier is not used
2016-08-02 12:50:46 2635 [Note] InnoDB: Compressed tables use zlib 1.2.3
2016-08-02 12:50:46 2635 [Note] InnoDB: Using Linux native AIO
2016-08-02 12:50:46 2635 [Note] InnoDB: Using CPU crc32 instructions
2016-08-02 12:50:46 2635 [Note] InnoDB: Initializing buffer pool, size = 128.0M
2016-08-02 12:50:46 2635 [Note] InnoDB: Completed initialization of buffer pool
2016-08-02 12:50:46 2635 [Note] InnoDB: Highest supported file format is Barracuda.
2016-08-02 12:50:46 2635 [Note] InnoDB: The log sequence numbers 2512357 and
2512357 in ibdata files do not match the log sequence number 2512396 in $
2016-08-02 12:50:46 2635 [Note] InnoDB: Database was not shutdown normally!
2016-08-02 12:50:46 2635 [Note] InnoDB: Starting crash recovery.
2016-08-02 12:50:46 2635 [Note] InnoDB: Reading tablespace information from the .ibd files...
2016-08-02 12:50:46 2635 [Note] InnoDB: Restoring possible half-written data pages
2016-08-02 12:50:46 2635 [Note] InnoDB: from the doublewrite buffer...
InnoDB: wrong number of columns in SYS_INDEXES record
InnoDB: wrong number of columns in SYS_INDEXES record
InnoDB: wrong number of columns in SYS_INDEXES record
InnoDB: wrong number of columns in SYS_INDEXES record
InnoDB: wrong number of columns in SYS_INDEXES record
17:50:46 UTC - mysqld got signal 11 ;
这似乎让我陷入了一个循环，我无法运行 mysql_upgrade 来升级数据库，但我无法成功重置 root 密码（尽管我 99.9% 确定我有正确的 root密码，它似乎不接受它。）
堆栈跟踪..
Thread pointer: 0x0
Attempting backtrace. You can use the following information to find out
where mysqld died. If you see no messages after this, something went
terribly wrong...
stack_bottom = 0 thread_stack 0x40000
/usr/sbin/mysqld(my_print_stacktrace+0x35)[0x8e2375]
/usr/sbin/mysqld(handle_fatal_signal+0x494)[0x666da4]
/lib64/libpthread.so.0[0x398620f790]
/usr/sbin/mysqld[0xac1017]
/usr/sbin/mysqld[0xac25fd]
/usr/sbin/mysqld[0xa3de77]
/usr/sbin/mysqld[0x9822dd]
/usr/sbin/mysqld(_Z24ha_initialize_handlertonP13st_plugin_int+0x48)[0x5aa958]
/usr/sbin/mysqld[0x6f01e1]
/usr/sbin/mysqld(_Z11plugin_initPiPPci+0xbb6)[0x6f4046]
/usr/sbin/mysqld[0x59ce38]
/usr/sbin/mysqld(_Z11mysqld_mainiPPc+0x425)[0x5a2055]
/lib64/libc.so.6(__libc_start_main+0xfd)[0x3985e1ed5d]
/usr/sbin/mysqld[0x593ba5]
有什么建议吗？谢谢
【问题讨论】：
-
信号 11 是一个段错误......这意味着你有某种内存访问冲突。也许是链接问题，错误地解决了包依赖关系？
标签： mysql